Methods, systems, and apparatuses, including computer programs encoded on a computer-readable storage medium for estimating the shape, size, and mass of fish are described. A pair of stereo cameras may be utilized to obtain right and left images of fish in a defined area. The right and left images may be processed, enhanced, and combined. Object detection may be used to detect and track a fish in images. A pose estimator may be used to determine key points and features of the detected fish. Based on the key points, a three-dimensional (3-D) model of the fish is generated that provides an estimate of the size and shape of the fish. A regression model or neural network model can be applied to the 3-D model to determine a likely weight of the fish.
Legal claims defining the scope of protection, as filed with the USPTO.
2. The method of claim 1, wherein the one or more swimming patterns associated with the type of fish include information identifying a swimming direction relative to a direction of water current associated with the type of fish, and the one or more cameras are positioned such that a longitudinal axis of a fish of the type of fish traveling along the swimming direction is substantially perpendicular to a view direction of the one or more cameras.
3. The method of claim 1, wherein the one or more swimming patterns associated with the type of fish include information identifying a depth at which the type of fish prefers to swim, and the one or more cameras are positioned at the corresponding depth.
7. The method of claim 1, wherein the information identifying the type of fish is received at the one or more computing devices in the form of a user-input.
13. The method of claim 11, wherein the one or more characteristics include at least one of: a weight, a size, or a length of the fish.
14. The method of claim 11, wherein a region of the fish pen captured in the second image is different from the region of the fish pen captured in the first image.
15. The method of claim 10, wherein the bounding box is generated via object detection by a trained recurrent convolutional neural network (RCNN), or semantic segmentation that segments the portion from a background.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 17, 2023
August 6, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.